VIC design session v50: reasoning

Lab note — VIC resonant sweep, 120 candidates, RANDLES+vapor-film ladder.

Honest flag first: this sweep was supposed to force each choke arm to 1.5–2.6mH (3.0–5.2mH total) to seat f_r in the 60–78kHz legal window against my 1.37nF cell. It did not. Every one of the top-five rows the scorer surfaced is the 0.1mH reference anchor sitting at f_r = 303.9kHz — 3× above the 1–100kHz practical band, hence the resonance VIOLATION flag on all five. So the intended high-mH rows either scored poorly or the scorer is over-rewarding the vapor-film term and ignoring the frequency penalty. I don't trust the ranking yet.

What the physics does show is consistent: the CGDE cold-chemistry window (3–6 eV, 80–250 Td) is comfortably lit here — 104–120 Td across the 0.1mm film at cell voltages of 262–301V, well under the 719V film Paschen. Displacement current stays 96.5%, breakdown margin ×1000+ on 2ppm water. And the classic trade-off is clean: raising choke turns (220→640) lifts inductance/copper but drops coil Q 322→111 and tank Q 25→22, so cell voltage and Td slide down together. More turns = more current restriction = less field = colder film.

I'm NOT adopting any of these five for build — they're out of band. Provisionally #1 as the physics reference (highest Td, best Q). Next session: re-run with the mH sweep actually applied, verify f_r lands 60–78kHz, and confirm the in-band rows still clear 100 Td. If they don't, the band and the window are in tension and I need to know that.